推荐权重1,触发词"anime pose, sketch, white background" 所有示例图均由T-anime-v4和nature footage v2生成。 第二张例图展示了这个lora生成的结果其实是很反ai的,用ai来对抗检测ai的ai。XD 要笑死了https://hivemoderation.com/ai-generated-content-detection 这是检测ai图的网站
recommend weight 1, trigger word "anime pose, sketch, white background" All examples are ganerated by T-anime-v4 and nature footage v2
The second example shows that the result generated by Lora is actually very anti AI, using AI to counter the AI that detects AI. XD funny
https://hivemoderation.com/ai-generated-content-detection here is the link
Description
FAQ
Comments (23)
This is excellent! As an artist who's pretty good at everything EXCEPT proportioning, this is my dream! A prime example of how AI can become a collaborative tool, rather than replacing humans outright as some suggest.
谢谢!
不用谢!thanks for your comment!
Agreed. I hope as this matures we can keep pushing for great shit like this, and not just more loras that just make everything into a big mush of the same look.
I totally agree with you, this is so helpful!
I love this alot too, im just very worried people are gonna use this to try to prove they are "real artist" and lie to the public for attention. It is very sad how such a wonderful tool is being soured by people like this, and most artist will continue to hate ai art as long as those people keep doing so
Hey, can you share the page where you verify that the image has been generated by AI or not, I was curious to try it
This is an amazing Lora, thank you for this
This lora is criminally underrated
Thanks for comment!
can you make a lora that bypasses hive ai detection
like a lora that bypasses them (general lora) I would be glad if you do 🤗
This is determined by the amount of data in the detection model. As long as I perform LORA on the data that is not labeled as AI generated in the detection model, the resulting graph will probably not be judged as AI
Im just wondering why you would want to do that, hopefully not to falsely claim yourself as an artist. Don't forget without artist's we would have have none of this technology and most of them are already pretty upset with how awful some people are trying to use it. Stuff like this only proves them right further. and the only way for artists to finally accept ai art is if the community starts to be a bit more respectful while using it. I personally use this to help me think of poses/ references. Most artists can also very easily tell if something is ai generated, whether a ai filter tells them or not. Most of them end up getting called out. Its very easy to tell apart from machine and art through another artists eye.. not so much for someone who does not have the eye of an artist so ofc those people think they will go unnoticed.
i dont really like using ai to make completed images, and instead like to use it as a helper for my art. this does that perfectly. i have mild aphantasia and cant really picture poses in my mind, so this kind of thing really helps my art. thank you so much.
I do this
Do you have any other models that do that, I'm looking for such models.
我想知道这是如何制作的 很棒的辅助工具 希望能有更多这种工具
正常训练一个泛化性高的lora就ok了
have you thought of making a Flux version?
maybe?
Would love a version for IL
<3
the idea of this is amazing








